This repository aim at parsing Wikipedia comparable corpora from linguatools and building bilingual co-occurence matrix for further analysis.
At this time, we only build en-zh and en-ja bilingual co-occurence matrix.
The execution steps are
- download corpora
- parse corpora by
Wiki_processing.ipynb
- apply NLP by
*_NLP.ipynb(.py)
- create co-occurence matrix by
*_NV_pairString.ipynb